[cap-talk] Singletons Considered Harmful
capibara at xs4all.nl
Tue Apr 6 22:57:46 PDT 2010
On Tue, April 6, 2010 22:07, David Wagner wrote:
> Kenton Varda wrote:
>> You're assuming that Leopold's
>> machine normally responds to requests in less than 1/100th of a second,
>> Leopold can influence this responsiveness within 1/100th second
>> and that Leopold's effects can be reliably distinguished from noise.
>> are all conceivable, but they are big "if"s, and I think a working
>> implementation would be very complicated.
> Yup, I'm assuming all that. I personally don't see them as big
> assumptions. 10 ms is an eternity in computer land. Error correcting
> codes might be effective at dealing with noise. But as you say, I
> have not implemented it, so I am just speculating, and what do I know?
This all seems pretty similar in concept to a setup that I created a proof
of concept many years ago. The usage of a simple (files only) single
public non-attacker owned web server on the Internet its file-system
caching timing as a covert channel. The difference measured by the
receiver of response time differences between requests of files that were
requested by the sender in the last few minutes versus files that were not
was way below the noise level.
There is a well known, and relatively easy to implement concept for
transmitting data over channels with a high level of noise, and that is
the usage of the autocorrelation properties of maximum length sequences.
Using these properties on already low bit-rate channels, your bit rates
won't impress anyone who doesn't understand that bit rates hardly matter.
In fact, at the time I did this, my communication skills were simply to
poor to convince anyone that my (if I remember correctly) approximately 3
bit a day bit rate could pose a real problem, or that my proof of concept
never set out to maximize the bit-rate :-(
More information about the cap-talk